The Lack of Correlation Between Economic Growth and Tax Rates

Another useful illustration of the point I made in this paper:

(source)

If you want to fight for lower tax rates, don’t say it’s because you want more economic growth. Say what you really mean.

What is Democracy? (67): The Form of Government That Offers the Best Protection Against Human Rights Violations

(source)

There is a clear correlation between the presence and quality of democratic government in a country and the level of respect for human rights in that country. That may sound obvious but it’s good to have some measured results. This paper for instance offers some clear evidence:

There is a substantial body of research devoted to understanding the relationship between democracy and government human rights performance. Most research centers on physical integrity rights but does not analyze the broader civil liberties encompassed by the category of “empowerment rights.” The dynamics of the relationship between the degree of democracy in a state and protection of empowerment rights might be different and improvements may take longer to emerge. This study examines the effects of democracy and democratic duration on empowerment rights scores, and it also uncovers time thresholds at which different scores are attained. The results show that regime type is more critical to the protection of empowerment rights than it is to physical integrity rights. Even in the earliest years of democracy there is a positive relationship between democracy and empowerment rights, but empowerment rights strengthen as countries gain democratic experience. …

Thus, countries with more institutionalized democratic regimes, as determined by the quality and longevity of democratic experience, are significantly more likely to respect both fundamental human rights and broader classes of civil liberties. … [A]lthough human rights protection is present in early years, it will usually be even greater after countries have had extended experience with democracy. (source)

Two graphs to back this up:

correlation between democracy and human rights

(source, click image to enlarge)

Here and here are other papers giving some further evidence.

The interesting thing about all this is not that there is a correlation – anyone following the news could have guessed as much. What we should care about are the reasons why there is a correlation. From the studies cited above we can see that the most important causal link is the one going from democracy to respect for human rights. In other words, there is a correlation because democracy causes respect for human rights. Vice versa may also be possible, although the argument is probably weaker. And then there may also be a hidden variable that can partially explain the correlation. For example, it may well be that prosperity and high GDP promote both democracy and human rights.

But then the next question is: how does democracy cause higher levels of respect for human rights? I guess this can happen in several ways:

• Democracies are more likely to be systems based on the rule of law and the rule of law is necessary for the protection of human rights.
• Democratic rulers know that they can’t get away with repression. They’ll be voted out if they try, or, worse, they’ll suffer the consequences of the rule of law, imposed on them by other branches of power in a system of checks and balances and separation of powers.
• Democracies have systems of judicial review which allow courts to void legislation that contradicts basic constitutional rights.
• Democracies have powerful non-violent mechanisms for dispute settlement, such as well-functioning courts. People don’t need to take the law into their own hands. Internal peace and limitations on violent behavior have beneficial effects on a number of human rights.
• Democracy is correlated with high levels of prosperity, and prosperity makes it easier to promote respect for human rights. Rights cost money.
• Democracies need human rights to function adequately (no democracy without free speech, free assembly, free association etc.), so they have an added incentive to respect them.

None of the above is meant to imply the following:

1. That we can delay the implementation of human rights norms in non-democratic states. Remember the remark at the beginning that the causal link probably goes in two opposite directions and that human rights can promote democratic government. After all, if people are allowed to express themselves, they will express themselves about the workings of their government, and that is the first step towards democracy.
2. That rights are never violated in democracies or never respected in non-democracies. It’s merely a matter of probability.
3. That there are no elements other than democracy that promote human rights. Of course there must be. I mentioned prosperity a moment ago. Democracy is not a sufficient condition, although probably a necessary one, at least in the long run, for the full set of human rights and for the equal enjoyment of all rights by all people.
4. That the beneficial effect of democracy on human rights is equal for all human rights or for all types of democracy. Well-developed and long-lasting democracies do better, as mentioned above, but perhaps also deep democracies, meaning democracies that provide a wide range of opportunities for democratic say.

More about the link between democracy and human rights here, here, here and here. More posts in this series are here.

Why Do We Need Human Rights? (36): The Economic Case Against Democracy

wealth in China

(source)

Democracy is a human right. But how do we justify this right? One common argument is that democracies tend to be wealthier than non-democracies. However, there’s some disagreement about this argument: not about the goodness of wealth and wealth-enhancing institutions, but about whether democracies are in fact such institutions. Impressive economic growth rates in non-democratic countries such as China have planted doubts in many people’s minds.

Some time ago, I offered a rather “philosophical” argument against the view that democracies perform worse economically than some types of authoritarian government (i.e. China-style). But in fact we’re dealing with empirically verifiable hypotheses here. So I looked for some numbers and found this article by Dani Rodrik:

The relationship between a nation’s politics and its economic prospects is one of the most fundamental – and most studied – subjects in all of social science. Which is better for economic growth – a strong guiding hand that is free from the pressure of political competition, or a plurality of competing interests that fosters openness to new ideas and new political players? …

Democracies not only out-perform dictatorships when it comes to long-term economic growth, but also outdo them in several other important respects. They provide much greater economic stability, measured by the ups and downs of the business cycle. They are better at adjusting to external economic shocks (such as terms-of-trade declines or sudden stops in capital inflows). They generate more investment in human capital – health and education. And they produce more equitable societies.

Authoritarian regimes, by contrast, ultimately produce economies that are as fragile as their political systems. Their economic potency, when it exists, rests on the strength of individual leaders, or on favorable but temporary circumstances. They cannot aspire to continued economic innovation or to global economic leadership. (source)

Some data on democracy and growth are here.

The darling of the “authoritarian=efficient” crowd is, of course, China. China has indeed performed extremely well economically under a rather authoritarian government. However, that government is much less authoritarian than it was during the post-WWII decades of stagnation and extreme poverty. So maybe it’s the relative move towards greater freedom that is the true cause of China’s economic performance, rather than its authoritarian government per se.

Moreover, China has done very well in terms of growth and poverty reduction, but in terms of levels of prosperity it’s still way behind most countries that are much more free. Its astounding progress is partly due to the very low starting point that was engineered by its authoritarian rulers.

And finally, the supposed economic success of authoritarianism in China – if it exists – isn’t necessarily proof of the economic ability of authoritarianism in general (authoritarian disaster stories are unfortunately far more common than authoritarian success stories). It may not even be proof of the economic ability of authoritarianism in China, since correlation doesn’t imply causation, especially not if there are only very few observations: China’s economic success may be due to other factors – and maybe this success would have been even greater without authoritarian government.

Jesus raises the dead man Lazarus back to the living world (mosaic from Ravenna, 500′s AD)

The economic case for authoritarianism is a bit like this: usually, people don’t return from the dead. But there’s this one guy, Lazarus, who did. Some claim that there was this other fellow, Jesus, who done the deed and made Lazarus walk again. There are no other Jesuses around, and this one Jesus only did his trick once. Nobody quite knows how he did it. Some say he just happened to be around when it occurred and people put one and one together. Lazarus would have walked anyway, perhaps even sooner had this other fellow not stolen all the attention.

More here, here, here and here on the myth of successful authoritarianism.

Related Articles

• The heavy hand (economist.com)
• Where will China’s long march end? (telegraph.co.uk)
• Is Democracy Necessary for Economic Success? (curiouscapitalist.blogs.time.com)
• India or China: Which Model for the Developing World? (curiouscapitalist.blogs.time.com)
• Democracy or Autocracy: Which is Better for Economic Growth? Becker (becker-posner-blog.com)

Statistical Jokes (45): Correlation Doesn’t Imply Causation, Ctd.

correlation v causation

(source)

More correlation-causation jokes. More statistical jokes.

Statistical Jokes (44): Correlation Doesn’t Imply Causation, Ctd.

(source)

This is a real study apparently, not a joke at all – at least not intended as one. And the “researcher” even goes out of his way to argue that correlation in this case does imply causation:

There was a close, significant linear correlation (r=0.791, P<0.0001) between chocolate consumption per capita and the number of Nobel laureates per 10 million persons in a total of 23 countries (Figure 1). When recalculated with the exclusion of Sweden, the correlation coefficient increased to 0.862. Switzerland was the top performer in terms of both the number of Nobel laureates and chocolate consumption. The slope of the regression line allows us to estimate that it would take about 0.4 kg of chocolate per capita per year to increase the number of Nobel laureates in a given country by 1. For the United States, that would amount to 125 million kg per year. The minimally effective chocolate dose seems to hover around 2 kg per year, and the dose–response curve reveals no apparent ceiling on the number of Nobel laureates at the highest chocolate-dose level of 11 kg per year. …

The principal finding of this study is a surprisingly powerful correlation between chocolate intake per capita and the number of Nobel laureates in various countries. Of course, a correlation between X and Y does not prove causation but indicates that either X influences Y, Y influences X, or X and Y are influenced by a common underlying mechanism. However, since chocolate consumption has been documented to improve cognitive function, it seems most likely that in a dose-dependent way, chocolate intake provides the abundant fertile ground needed for the sprouting of Nobel laureates. (source)

I still have a hard time believing this is for real.

More correlation-causation jokes. More statistical jokes.

Statistical Jokes (39): Correlation Doesn’t Imply Causation

(source)

(source)

More correlation-causation jokes here, here, here, here, here and here. More on homicide rates here. More statistical jokes here.

Human Rights Maps (157): Homicide in NYC is Primarily a Problem of and for Male African Americans

Apparently, it’s more dangerous to be a male black person in NYC than a person of any other race or gender:

(source, where you can find an interactive version of these maps)

African Americans represent only 25% of NYCs population, but 61% of murder victims. The racial distribution of the perpetrators is strikingly similar to the racial distribution of the victims; and men are not only the main victims but also the perpetrators in 92% of cases.

A note of caution: correlation doesn’t imply causation. In this case, this means that the race of most of the perpetrators shouldn’t lead you to the conclusion that black people are more likely to engage in murder because they are black. A third element, hidden in the correlation and more common among blacks, is most probably the cause of the high murder rate (perhaps poverty). In which case, distorted homicide rates may be a symptom of racism and discrimination.

Another note of caution: a common feature of a lot of statistical data in map form is that they exaggerate the prevalence of the phenomenon that is measured, and so it is with these images of murder in NYC. The town is full of it, if you can believe the images. But that’s obviously not true. 500 or so homicides per year, on a total population of 8 million, amounts to one murder per 16.000 people, only slightly higher than the 1 in 18.000 for the US nationwide (it’s not surprising that it’s higher for a densely populated urban area).

Also, the numbers have trended downwards in NYC:

(source)

Apparently the same pattern can be seen in Chicago:

(source)

And Washington DC as well – data are here:

(source, where you can find an interactive version)

More maps on violence are here, and more human rights maps in general are here.

Human Rights Nonsense (29): The Link Between Porn and Terrorism

scene from The Good Old Naughty Days

(source)

Pornography is not a necessary cause of terrorism. The abolition of pornography would not lead to the cessation of terrorism in the world. Terrorism existed well before graphic pornography and its mass spread via the internet.

Likewise, pornography is not a sufficient cause for terrorism. There are pornography users, even addicts, who do not become terrorists. Given how widespread the viewing of pornography is today, if the direct result of each individual’s pornography use were terrorist violence, one could conceivably argue that pornography proliferation would pose a more widespread threat to human existence than nuclear proliferation.

Yet pornography now appears frequently in the possession of violent terrorists and their supporters, including Osama bin Laden. …

I wonder whether the pornography of today—now ubiquitous and increasingly grotesque—is one of the influences warping the mentality of those who aspire to or who actually go on to engage in ever more grotesque public violence. … Why, after all, would an al-Qaeda affiliate, as reported in 2009 from interrogations in Mauritania, select pornography to target new recruits? We need to know.

As terrorism researchers Daniel Bynum and Christine Fair point out in an article about the modern terrorists we have been pursuing, especially since 9/11, the fact of the matter is that “they get intimate with cows and donkeys. Our terrorist enemies trade on the perception that they’re well trained and religiously devout, but in fact, many are fools and perverts who are far less organized and sophisticated than we imagine. Can being more realistic about who our foes actually are help us stop the truly dangerous ones?” (source)

Yes, indeed, “we need to know”. Perhaps. Or perhaps there is nothing to know. Who knows? I have rarely seen a pile of insinuations so completely devoid of data and evidence. I do admit that the effects of porn consumption on people’s actions are a worthy subject of scientific investigation. Some forms of pornography can have a dehumanizing effect and may change men’s perceptions of women, perhaps to such an extent that porn can lead to violent acts such as rape. But the evidence available so far is mixed. And in the specific case of terrorism caused by porn, all we have are flimsy anecdotes and insinuations. I’m sure you can find just as many little stories about terrorists and violent games, terrorists and early child abuse, terrorists and poverty, terrorists and beards and so on.

The story above is just a free floating riff. “The U.S. government has had opportunity to observe, and in many cases, acquire, personal media from untold numbers of those involved in terrorism and the support of terrorism … [we] may be sitting on a massive data set for studying the intersection of pornography use and support for twisted violence such as terrorism” [my emphasis]. But then again, we may not be sitting on a massive data set. However, that’s no reason not to speculate, right? As I see it, there isn’t even a correlation, let alone evidence of causation. Just random anecdotes that are of no help at all explaining terrorism. You need to do better than that if you want to find the causes of some of today’s most horrific human rights violations.

More posts in this series are here.

Children’s Rights (13): Minimum Age of Marriage Laws Reduce Incidence of Child Marriage

(source)

In many countries, it’s customary for girls to marry at a very young age, voluntarily or not. This practice is detrimental to the human rights of women, as I argued before.

In the developing world, more than one third of women aged 20 to 24 report that they were married or in a union by the age of 18. (source)

This practice is often legally entrenched:

In 50 countries, the minimum legal age of marriage is lower for females. (source)

However, it seems that the law can also work the other way:

(source, click image to enlarge)

Or perhaps the causation goes the other way: countries where customs are against early marriage also adopt laws stipulating a high minimum marriage age. In general, we shouldn’t be too optimistic about the power of legislation.

More on child marriage here. More human rights facts here.

Statistical Jokes (32): Correlation and the Direction of Causation

(source)

More on correlation and causation. More statistical jokes.

Why Do Countries Become/Remain Democracies? Or Don’t? (20): Education Again

It’s a common assumption that democracy is driven by levels of education:

• Less educated people are – supposedly – easier to oppress and more willing to accept extreme and simplistic ideologies that authoritarian rulers can exploit. They are also said to be less tolerant, and therefore less willing to accept freedoms and rights that protect outgroups.
• Once people become more educated, they start earning more. And because they earn more, they have more leisure time. And because they have more leisure time, they have more opportunities to engage in various activities. And because they have these opportunities, they start to demand the freedoms they need to take up these opportunities. Better education itself, irrespective of the higher earning potential that goes with it, opens up opportunities to do things, and hence drives the demand for the freedom necessary to do things.
• More educated people are also more aware of the ways in which their governments oppress them and of the liberties enjoyed in other countries, and they are better able to organize and mobilize against their governments.
• Maslow’s theory about the hierarchy of needs also plays a part: when lower needs – such as food, clothing and shelter – are met, then the preconditions are fulfilled for the appearance of higher needs. Higher education levels, because they help to fulfill lower needs, assist the appearance of needs such as self-actualization, self-esteem and belonging, needs that require freedom for their realization.
• Democracy requires a certain level of education among citizens in order to function properly. Of course, it’s not because B requires A that A results in B; claiming that education results in democracy because democracy needs education would mean committing a logical error. However, the fact that democracy needs education does probably increase the likelihood that democracy will follow from more education. At least the absence of some level of education will diminish the chances of democracy.
• And, finally, more education improves the capacity to make rational choices, and democracy is essentially a system of choice. Democracy will therefore intrinsically appeal to the higher educated.

And indeed, there is a correlation – albeit not a very strong one – between levels of education and degrees of democracy:

education and democracy correlation

(source)

The correlation may be due to the fact that democracies are better educators, but there are some reasons to believe that part of the causation at least goes the other way. Anecdotal evidence is provided by the recent Arab Spring: education levels in Arab countries have risen sharply in recent decades.

More posts in this series are here.

(image source)

What is Democracy? (55): A System For Signaling Disapproval of the Economy

statistics cartoon by R. Crumb

(source)

“It’s the economy, stupid“. The famous phrase suggests that economic basics rather than social or cultural issues, politicians’ personal merit, foreign policy successes etc. determine democratic outcomes. People vote against incumbents when unemployment is high and GDP growth low, whatever the causes of the economic downturn. One can accuse George H.W. Bush of many things but he wasn’t by far the sole or main cause of the recession that propelled Clinton to power.

It seems that people use democratic elections – especially high profile one such as presidential elections – to signal disapproval of the economy, whatever the real responsibility of individual politicians for the state of the economy (it’s silly to assume that individual politicians, even American presidents, have the power to dramatically change the unemployment rate, for better or worse).

This graph shows a clear correlation between incumbent margins of victory in US presidential elections and changes in the unemployment rate:

incumbent margin of victory vs change in unemployment rate

(source, incumbents running for a second full term won when the unemployment rate was falling and they lost when it was going up)

The unemployment rate in 1980 was 7.2%, up considerably from the year before; Jimmy Carter duly lost his bid for reelection. In 1984, the unemployment rate was actually higher at 7.5%, but was on its way down from the previous year; Ronald Reagan was reelected in a landslide. (source)

Granted, the number of observations is low, too low to be certain about the correlation (and others have expressed doubts about the data).

Some more certainty is given by the fact that it’s not just unemployment but also income that is correlated with election results:

(source)

If we assume that the economy does indeed determine democratic outcomes in this way, then we face a problem because some of the traditional justifications of democracy become unavailable. Democracy is supposed to improve the quality of politicians: when a politician has to face popular judgment and has to pass the test of accountability, she will try harder to respect the will of the people, to avoid engaging in corruption and to generally do a good job, because doing so will convince the people that she deserves reelection. Also the freedom of the press is in part justified on this basis. A free press is able to give people the information about politicians necessary to make an informed judgment at the next election. Government transparency combined with accountability as a prerequisite for reelection provides politicians with the necessary incentives to do a good job, or at least a job that is considered good by an informed majority of public opinion.

If, however, democratic election results are determined less by the actual way in which politicians do their job than by the economic basics, then we lose this justification. If a politician won’t be reelected during an economic downturn, even if she does all that’s humanly possible to avoid it or lessen its impact, she has one less reason to do her best. She may still do her best out of a sense of responsibility or public service and remain unconcerned by electoral prospects, but we should not underestimate the motivation provided by the mere fact of having and retaining a position of power.

More posts in this series are here. More on signaling here, and on accountability here.

Why Do Countries Become/Remain Democracies? Or Don’t? (19): Psychological Reactions to the Threat of Disease

microscope

(source)

There sure are many reasons why countries become or fail to become democracies. In this blog series I’ve mentioned climate, geography, inequality, external triggers, prosperity, religion, resources, education etc. An original approach to this question looks at psychological reactions to the threat of disease:

Conventional explanations for a country’s political system would draw on its history, economy and culture. Randy Thornhill from the University of New Mexico, Albuquerque, however, thinks it might be determined by the threat of disease in a region. This triggers psychological biases, which originally evolved to prevent illness spreading, that also hinder the emergence of democratic ideals. (source)

The logic is that people develop psychological reactions – call them biases – which they need to protect themselves against infectious diseases, and these reactions in turn make it difficult to adopt democracy, individualism and an attitude of criticism of authority.

The starting point for Thornhill and Fincher’s thinking is a basic human survival instinct: the desire to avoid illness. In a region where disease is rife, they argue, fear of contagion may cause people to avoid outsiders, who may be carrying a strain of infection to which they have no immunity. Such a mindset would tend to make a community as a whole xenophobic, and might also discourage interaction between the various groups within a society – the social classes, for instance – to prevent unnecessary contact that might spread disease.

What is more, Thornhill and Fincher argue, it could encourage people to conform to social norms and to respect authority, since adventurous behaviour may flout rules of conduct set in place to prevent contamination. Taken together, these attitudes would discourage the rich and influential from sharing their wealth and power with those around them, and inhibit the rest of the population from going against the status quo and questioning the authority of those above them. This is clearly not a situation conducive to democracy. (source, source)

What is, initially useful for public health, becomes detrimental for self-government:

[S]pecific behavioural manifestations of collectivism (e.g. ethnocentrism, conformity) can inhibit the transmission of pathogens; and so we hypothesize that collectivism (compared with individualism) will more often characterize cultures in regions that have historically had higher prevalence of pathogens. Drawing on epidemiological data and the findings of worldwide cross-national surveys of individualism/collectivism, our results support this hypothesis: the regional prevalence of pathogens has a strong positive correlation with cultural indicators of collectivism and a strong negative correlation with individualism. (source)

(source, dots represent countries)

Statistical Jokes (29): Causation-Correlation Again

(source)

Here’s a similar one:

(source)

More on the causation-correlation problem here. More statistical jokes here.

Lies, Damned Lies, and Statistics (35): Sample Sizes, Ctd.

cherry picking

(source)

This isn’t the first time I mention sample sizes as a common problem in statistics. Usually, the problem is one of survey design: insufficiently large sample sizes for respondents produce unreliable survey results.

However, the same error – or fraud, when the error is willful – can occur in data interpretation. Take a look at this graph by John Taylor:

(source)

The problem?

Taylor’s conclusion: The data on spending shares show that the most effective way to reduce unemployment is to raise investment as a share of GDP. But why begin the scatter plot in 1990? There’s no good reason. In fact, most folks typically download the entire history of available macro data. … The chart below goes back to 1948:

(source)

This is a form of cherry-picking data that allows you to “prove” a strong correlation where there’s actually none at all. In this way, you’ll find a correlation in almost all data sets, as long as you pick a sufficiently small sample of the set. In this example, you can only limit the selection to the last two decades if you have a good argument about why the economy is different now compared to some decades ago, and why there’s a correlation now when there wasn’t before. However, that argument – which would be interesting – seems to be lacking. And if it’s lacking,  there’s no excuse for cherry picking the last two decades.

Other examples of cherry-picking are here. More posts about lies and errors in statistics are here.

Lies, Damned Lies, and Statistics (33): The Omitted Variable Bias, Ctd.

(source, click image to enlarge)

I discussed the so-called Omitted Variable Bias before on this blog (here and here). So I suppose I can mention this other example: guess what is the correlation, on a country level, between per capita smoking rates and life expectancy rates? High smoking rates equal low life expectancy rates, right? And vice versa?

Actually, and surprisingly, the correlation goes the other way: the higher smoking rates – the more people smoke in a certain country – the longer the citizens of that country live, on average.

Why is that the case? Smoking is unhealthy and should therefore make life shorter, on average. However, people in rich countries smoke more; in poor countries they can’t afford it. And people in rich countries live longer. But they obviously don’t live longer because they smoke more but because of the simple fact they have the good luck to live in a rich country, which tends to be a country with better healthcare and the lot. If they would smoke less they would live even longer.

Why is this important? Not because I’m particularly interested in smoking rates (although I am interested in life expectancy). It’s important because it shows how easily we are fooled by simple correlations, how we imagine what correlations should be like, and how we can’t see beyond the two elements of a correlation when we’re confronted with one that goes against our intuitions. We usually assume that, in a correlation, one element should cause the other. And apart from the common mistake of switching the direction of the causation, we often forget that there can be a third element causing the two elements in the correlation (in this example, the prosperity of a country causing both high smoking rates and high life expectancy), rather than one element in the correlation causing the other.

More posts in this series are here.

Migration and Human Rights (34): The “Criminal Immigrant” Stereotype, Ctd.

I presented some data debunking the criminal immigrant stereotype a few times already. It’s simply not true that immigration leads to an increase in crime rates. True, immigrants are often – but not always – relatively poor, undereducated and – initially at least – not well adjusted to their host community. But none of that seems to be a sufficient reason for higher crime rates among immigrants.

On the contrary, there’s some evidence here of immigration actually reducing crime rates:

During the 1990s, immigration reached record highs and crime rates fell more precipitously than at any time in U.S. history. And cities with the largest increases in immigration between 1990 and 2000 experienced the largest decreases in rates of homicide and robbery. … Wadsworth contends that looking at crime statistics at a single point in time can’t explain the cause of crime rates.

Using such snapshots in time, Wadsworth finds that cities with larger foreign-born and new-immigrant populations do have higher rates of violent crime. But many factors—including economic conditions—influence crime rates.

If higher rates of immigration were boosting crime rates, one would expect long-term studies to show crime rising and falling over time with the influx and exodus of immigrants. Instead, Wadsworth found the opposite. (source)

There’s yet another study here showing that Hispanic Americans are less violent than whites or blacks.

A simple juxtaposition of immigration trends and crime trends can already make clear how silly it is to claim that higher immigration rates produce higher crime rates:

(source)

Property crime rates in the United States (1986-2005) Source: FBI Uniform Crime Reports

What could be the explanation? Why does immigration reduce crime rates? Maybe the culture and religion of the immigrants has something to do with it. Or maybe it’s true that people migrate because they want to have a better life, and that engaging in crime is incompatible with this motivation. Or perhaps the fact that immigrants tend to live in extended families and close-knit communities discourages crime.

I’ve said it before: although correlation doesn’t always equal causation, these numbers are compelling, even if we accept some possible caveats (illegal immigrants, when committing a crime, are perhaps more likely to flee abroad and hence not end up in incarceration statistics, and there may be some underreporting of crime in communities with a lot of illegal immigrants). Politicians should therefore stop exploiting irrational fears about immigrant crime for their own partisan gain. You don’t solve the crime problem by closing the border, and certainly not by ignoring overwhelming scientific evidence.

Related Articles

• Massive Drop in Crime Accompanies Major Wave of Immigration (criminaljustice.change.org)
• Fox: Confusion on illegal immigrants, crime (boston.com)
• New study on challenges ‘immigrant criminality’ theory (thegrio.com)
• more illegal immigrants does NOT = more violent crime (schansblog.blogspot.com)

Terrorism and Human Rights (29): Terrorism, Caused by Poverty or Repression?

This is a follow up from some previous posts (here, here and here) claiming that poverty doesn’t cause terrorism, at least not usually, and another one (here) claiming that rights violations are a better predictor (and, conversely, respect for human rights predict reduced terrorism). I’ve found this paper (gated unfortunately) supporting those claims.

The empirical results reported here show that terrorist risk is not significantly higher for poorer countries, once the effects of other country-specific characteristics, such as the level of political freedom, are taken into account. … lack of political freedom is shown to explain terrorism, and it does so in a nonmonotonic way. Countries with intermediate levels of political freedom are shown to be more prone to terrorism than countries with high levels of political freedom or countries with highly authoritarian regimes. …

On the one hand, the repressive practices commonly adopted by autocratic regimes to eliminate political dissent may help keep terrorism at bay. On the other hand, intermediate levels of political freedom are often experienced during times of political transitions, when governments are weak, and political instability is elevated, so conditions are favorable for the appearance of terrorism. (source)

More posts in this “human rights facts” series are here. More on terrorism and on poverty.

Why Do Countries Become/Remain Democracies? Or Don’t? (13): Prosperity

I already mentioned in a previous post how democracy is correlated with prosperity. There’s a much higher proportion of democracies among rich countries than among poor countries. The level of national income is the most important factor explaining inter-country variations in the degree of democracy. If we assume from this correlation that there is a causal link from prosperity to democracy, then low income is the most important barrier to democracy. But the causal link probably goes in both directions. Countries aren’t just democratic – or remain so – because they prosper (among other reasons), but it’s also the case that countries prosper to some extent because they are democratic (disproving the often heard claim that economic development requires authoritarian government). I gave an overview of the reasons why prosperity promotes democracy here, and why the opposite is also the case here.

The correlation between democracy and prosperity is obvious from this graph (at least for non-Muslim countries):

(source)

The stronger one of the causal links seems to be the one going from prosperity to democracy rather than vice versa. If you accept that, there’s an additional question (it’s one made famous by Przeworski and Limongi): are there more democracies among rich countries than among poor countries

• because economic development increases the likelihood that countries will undergo a transition to democracy (this is often called modernization theory), or
• because economic development makes democracies less likely to fall back into dictatorship?

Przeworski and Limongi found that affluence makes it very unlikely that a shift from democracy to dictatorship occurs, while Boix and Stokes find that there is an effect of affluence on the likelihood of a shift to democracy. Both effects are visible in this graph:

(source)

It’s likely that the economic effect on transition towards democracy is a bit smaller than the effect halting the opposite transition. The reason is probably the fact that the transition from democracy to authoritarianism is in se much easier than the other way around. Some even say that democracy is inherently suicidal. Whatever the merits of that claim, it’s obvious that an authoritarian leader has the resources and the necessary lack of scruples to cling to power. Especially when his country becomes more prosperous. He can then use this prosperity to bribe the population into submission, and buy the arms and security forces when this doesn’t work.

Again, economic development isn’t a sufficient or even necessary prerequisite for democracy to appear or to survive. Things are more complicated than that and many other factors are in play, including conscious human activity and volition. People can decide to make or destroy a democracy at any level of economic development.

More posts in this series are here.

The Ethics of Human Rights (29): Should Taxation Be a Tool For Economic Efficiency or For Social Justice?

(source, source)

Taxation is a recurring theme in political discussions between people of the left and right. People of the left see taxation as a tool for social justice. They tend to prefer rather high taxation rates and a progressive taxation system:

• High taxation rates bring in revenues that are large enough to enable the government to spend on programs and transfers that are designed to promote social justice: unemployment benefits, poverty reduction policies, education, healthcare etc.
• Progressive taxation rates are just because they impose relatively (and not just absolutely) higher taxes on people who are more able to pay, and, in addition, reduce income inequality and hence realize another goal of social justice.

People on the right usually favor low tax rates and a non-progressive taxation system (either a proportional system in which everyone pays the same share of their income, or a regressive system in which everyone pays more or less the same amount in taxes). Rather than on social justice, they focus on the economic effects of taxation.

• They reject high taxation rates because they claim that these high rates discourage people and are a disincentive to hard work, effort and investment. Because high rates limit effort and investment, they also limit productivity, innovation, international competitiveness and job creation.
• They also reject progressive tax rates because high tax rates for high incomes discourage those people who work relatively hard (they work hard supposedly because they earn a lot) and who are most likely to innovate, to be productive and competitive and to create jobs.
• However, they don’t necessarily favor regressive taxes because they are equally hostile to high tax rates for low income people, albeit for other reasons. High taxes for low income people discourage them from entering the labor market and hence inflate unemployment. Still, they claim that the worst damage is done by high taxes on the higher incomes, which is the reason they reserve particular scorn for progressive taxation systems. Because high tax rates for the wealthy punish the most productive elements in a society, the whole of society suffers. More productive people will limit their productivity because they don’t want to fall into a higher tax bracket, and the money they pay in taxes can’t be invested in the economy. High tax rates, especially for the rich, have an unacceptable cost in terms of economic efficiency. Keeping taxes low, on the contrary, and allowing wealthy people to use their money in the economy, will ultimately benefit everyone (this is the so-called Trickle-Down theory).

Of course, this distinction between left and right is a caricature. Most people on the left are also concerned about economic efficiency, and most on the right are not insensitive to questions of social justice. The extremes are hardly ever encountered in real life: no one wants to limit taxes to such an extent that economic efficiency is promoted but no money is left for justice, and no one wants to put tax rates at such a high level that there is ultimately no more economy to tax. (The latter concern is expressed in the famous Laffer Curve arguing that beyond a certain level of tax rates government revenues in fact decrease instead of increase. At very high rates there is no longer any incentive for a rational taxpayer to earn any income and hence tax revenues will decline while tax rates increase. However, it isn’t clear what “very” in the previous sentence actually means and where exactly the tipping point is situated).

Graphically, we can represent this in the following image:

Normal political discourse takes place in the light-gray area.

Personally, I believe that the concerns of both right and left are justified and need to be balanced, and that too much focus on either the element of efficiency or justice is detrimental to the other element. On the one hand, there’s only so much money a government can raise without wrecking the economy, and justice isn’t only about spending money (there can even be perverse effects such as unemployment traps, welfare dependency etc.). On the other hand, there’s only so much an efficient economy can do to realize social justice all by itself and quasi-automatically (remember the invisible hand…). To quote Matthew Yglesias’ sarcastic comment on the skyrocketing incomes of the U.S. top 400 earners in the decades leading up to the 2009 recession:

As is well-known, the Top 400 are considerably more talented than the rest of us. And [the] decline in their tax rates has created exciting new incentives for them to apply their talents. And that, in turn, is why the 2000s were a so much more economically successful decade than the 1990s, not just for the Top 400 but for the rest of us as well. Thanks to their skyrocketing incomes and falling tax rates, we’re currently [during the 2008-2009 recession, FS] all enjoying the fruits of prosperity, rapid growth, and low unemployment. Thanks rich guys! (source)

A similar sentiment is expressed in this clip from the Daily Show (I’m unable to embed it; skip to the 4th minute or so).

Here’s one very specific example of the way in which taxation can promote social justice:

Again, personally, if I lived in the U.S., I would probably be on the left side of the arrow in the efficiency v justice graph above, since I believe taxes in the U.S. are relatively low and can be raised without too much harm to economic efficiency. The resulting government revenues could then be spent on improving the social safety net and promoting social justice. It’s difficult to imagine for a European that a country such as the U.S. doesn’t offer health insurance to millions of its citizens. Also, unemployment benefits are quite stingy in the U.S., both in terms of eligibility and duration: only one third of the unemployed qualify for benefits and only for 26 weeks (extendable during recessions if the Republicans don’t object, as they infamously did beginning of 2010):

share of unemployed receiving benefits in the US

(source)

The system of unemployment benefits could easily be improved without perverse effects or harm to economic efficiency. And there are other areas of possible improvement as well.

However, as a European in Europe, I think I’m probably more to the right of the graph since there’s a strong argument that the social safety net in Europe (at least in some countries) has harmed European competitiveness, labor market participation and innovation.

Still, is there evidence of this? What do the data say about high tax rates harming economic efficiency, in Europe and in general? Is the conservative case against taxes as strong as it seems? I’m afraid not. In this previous post, I already presented some evidence that the effect of reasonably rather than extremely high rates on economic efficiency is minimal at best. I now present some more evidence from Lane Kenworthy about the U.S. and other affluent countries (always keeping in mind that correlation doesn’t imply causation and that the absence of a large negative effect of high taxes doesn’t preclude the possibility that lower taxes would have had a large positive effect). One measure of economic efficiency is economic growth. If we plot economic growth rates for the U.S. against tax rates for the wealthy we get the following picture:

effect of tax rates on economic growth in the US

(source)

If anything, higher tax rates lead to more growth. But of course there can be catch-up effect: higher rates producing their effects only years later. That’s taken into account in the following graphs, which also show that an international comparison doesn’t prove that countries with higher tax rates have lower growth:

effect of tax rates on economic growth international comparison

(source)

If we have a look at the data about the effect of high tax rates on unemployment (another conservative concern), we also see that we shouldn’t panic about taxes:

effect of tax rates on unemployment international comparison

(source)

Now, if there is no good reason not to tax at a moderately high level, based on concerns about economic efficiency, the question remains whether there is a good reason to tax based on social justice reasons. Given the caveat that social justice isn’t all about government spending (I argued here that it is primarily about something else) and that such spending can in some cases have perverse effects (see above), I do believe that some spending is necessary in some cases, and that relatively high tax rates are necessary to produce the revenues required for this spending.

Again following Kenworthy, I believe that relatively high tax rates are acceptable and even necessary to create the revenues required for social justice policies, but that progressive tax rates in themselves don’t do the job of reducing income inequality, contrary to what is often claimed as a justification for progressive rates. That doesn’t mean that we shouldn’t reduce income inequality (it’s quite high in the U.S.) – there are good reasons to try. It just means that progressive taxation in itself won’t do the job. The important thing is to have high tax revenues which can then be spent in transfers and services that reduce income inequality and achieve other goals of social justice. Yet, I still think a progressive system is required, not because of its supposed effects but simply because it is just in itself, compared to proportional or regressive systems. A person with more income can afford to pay, not merely more in an absolute sense but more in the sense of a larger share of his or her income.

More posts in this series are here.

Why Do Countries Become/Remain Democracies? Or Don’t? (11): The Relative Cost of Freedom and Dictatorship

When dictatorial governments come under international pressure to improve the human rights situation in their countries, they often react by stating that they govern developing countries and don’t have the resources that are necessary to make improvements. Such statements have some plausibility. A judiciary, a well-trained police force, a functioning system of political representation etc. all require funding. (I pointed this out some time ago in a post about negative and positive human rights).

However, to some extent this explanation is no more than an excuse: you don’t need money to stop persecution of dissidents, to lift restrictions on the media, to allow demonstrations etc. On the contrary, you save money by doing so. You don’t need a large police force or paramilitary force; you don’t need strong government controls of every aspect of society and the economy; you don’t need to bride your citizens into acceptance of the state etc. But obviously the goal of dictators isn’t to save money and make the country better off by investing that money in the economy.

On the other hand, it remains true that the adequate defense of freedom, rights and democracy requires money, which is probably why rich countries usually score higher in freedom indexes (see also here). And, consequently, governments can save money by limiting freedom and by oppressing people.

So, both oppression and freedom cost money, and both a reduction of oppression and a reduction freedom save money. The question is then: what is, overall, the cheapest? A dictatorship or a democracy? And how can we know? Well, one possible indicator could be government spending as a percentage of GDP. If democracies have a systematically higher percentage, one could say that freedom costs more than oppression (on the condition that there isn’t a third variable explaining why democracies spend more).

However, one look at the data tells you that there isn’t much of a correlation between freedom and government spending, or between oppression and government spending. There are some countries that oppress a lot with not a lot of money – “not a lot” in relative terms compared to GDP. China and Saudi Arabia for example. And there are others that do need a lot of money (a large share of the economy) to keep the bosses in place. Cuba and Zimbabwe for example. But perhaps that is because their GDP is so low, not because they need a lot of money to oppress. In other cases, such as Saudi Arabia we may think they don’t spend a lot on oppression but we are fooled because their GDP is relatively high. And anyway, even dictatorships use some part of their state budget for things that aren’t quite so bad.

Likewise for freedom: freedom comes “cheap” in the U.S., and is “expensive” in Sweden. Between quotation marks because government spending over GDP is a very imprecise measure of the cost of freedom or oppression, for the reasons just given. It’s not because a country’s GDP doubles thanks to higher oil prices that the cost of freedom also doubles. Freedom (like oppression) costs money but not money as a fixed percentage of GDP.

Alternatively, you can also look at the tax burden. Here, the data show that countries that impose the highest taxes are also the ones that are most free (Scandinavia obviously ranks high on both accounts). But is that because freedom costs so much more than oppression? Perhaps the answer is “yes” if you include in “freedom” the things that make freedom possible, such as good healthcare, education etc.

But perhaps a more interesting and useful question would be: what cost considerations or economic incentives would produce a move towards democracy or away from democracy? It’s clear that a crisis of some sort – 9/11, a war, or, more appropriately in the current context, an economic recession or depression (see the Roosevelt cartoon below) – encourages democratic leaders to abridge certain rights, freedoms and democratic procedures. In the case of an economic crisis, the claim is that freedom and proper democratic procedures are just too expensive economically. A swift resolution of the crisis requires strong centralized intervention.

It’s also widely accepted that one of the causes of the demise of the Soviet Union was the unbearable cost of oppression. I think it’s better foreign policy to try to make oppression as costly as possible, rather than trying to make freedom as cheap as possible. Freedom tends not to be very cheap, I guess. And when it is, it’s probably not really freedom.

More about Putin. More on the determinants of democracy.

Migration and Human Rights (26): The “Criminal Immigrant” Stereotype, Ctd.

Contrary to right-wing rhetoric and popular belief (examples here and here), there isn’t much of a correlation between Latino immigration in the U.S. and crime rates. That’s something I discussed before, but I want to revisit the subject because there’s an interesting new article about it here confirming my claims (to make it even more interesting: it’s from a conservative magazine).

Nearly all of the most heavily Latino cities have low or even extremely low crime rates, and virtually none have rates much above the national average. Eighty percent Latino El Paso has the lowest homicide and robbery rates of any major city in the continental United States. This is not what we would expect to find if Hispanics had crime rates far higher than whites. Individual cities may certainly have anomalously low crime rates for a variety of reasons, but the overall trend of crime rates compared to ethnicity seems unmistakable.

Maybe we should assume that the numbers are bit too rosy because of the tendency of illegal immigrants to underreport crime (although the article tries to correct for underreporting by comparing homicides – almost no underreporting – to overall crime). Also, the likelihood of underreporting by illegal immigrants can be offset by a possibly equal effect of criminal restraint on the part of illegal immigrants: for the same reasons that they underreport crime – fear of contacting the authorities and being identified as illegal immigrants – they stay out of trouble with the police and try to act decently.

However, if we look at it from another side, we see that incarceration data show somewhat higher levels for Hispanics or immigrants (although most Hispanics are American-born, the vast majority still comes from a relatively recent immigrant background):

the age-adjusted Hispanic incarceration rate is somewhat above the white rate—perhaps 15 percent higher on average. (source)

Still, one can’t simply conclude from this that crime is more rampant among Hispanics or immigrants. It’s still possible that instead of higher criminality we simply witness the result of harsher treatment of those sections of the population by the judicial system. Also, incarceration rates are inflated because many immigrants are in jail not because of ordinary crimes, but because of infractions of immigration law; you should exclude the latter if you want to compare Hispanic and white criminality (unless you consider infractions of immigration law as essentially equivalent to ordinary crime, which is not altogether insane; but the point of this post is to examine the claim that there are more ordinary criminals among Hispanic immigrants than among [longtime] citizens).

In addition, you should correct incarceration rates for age and gender: in general, most criminals are young men, and it happens to be the case that most immigrants are also young men. So the likelihood that immigrants end up in prison is – slightly – higher compared to the general population, not because they’re Hispanics but because they are young men. Any other, non-immigration related influx of young men in a certain area – e.g. military demobilization or a huge construction project – would have an effect on crime. (If you don’t correct for this, you’re making a common statistical mistake: see here for other examples of the “omitted variable bias”).

Finally, immigrants are relatively poor and there is a link between poverty and crime. So that can also explain the higher incarceration rate for immigrants. If you link the higher probability of poor people engaging in crime with the fact that poor people have lower quality legal representation, you have a double explanation. So, again, if Hispanics do end up in jail more often, perhaps it’s because they’re relatively poor, not because they are Hispanics and somehow racially prone to crime.

All this is limited to the U.S. People can still make the case that immigration in other countries promotes crime, but that case is made harder by the false claims about the U.S. (At least in France there’s no proof of the share of immigrants in the population having a significant impact on crime rates). These false claims are always based on anecdotes, and you’ll always be able to find criminals with foreign sounding names in order to whip up a frenzy against immigration, thereby satisfying your racist hunger and building a political following of ill-informed voters. Again a clear demonstration of the usefulness of statistical analysis in human rights issues and the danger of anecdotal reasoning.

Bonus paper here. Quote:

We examine whether the improvement in immigrants’ relative incarceration rates over the last three decades is linked to increased deportation, immigrant self-selection, or deterrence. Our evidence suggests that deportation does not drive the results. Rather, the process of migration selects individuals who either have lower criminal propensities or are more responsive to deterrent effects than the average native. Immigrants who were already in the country reduced their relative institutionalization probability over the decades; and the newly arrived immigrants in the 1980s and 1990s seem to be particularly unlikely to be involved in criminal activity.

More on migration.

Measuring Poverty (5b): Poverty in Africa

Here‘s an interesting paper by Sala-i-Martin and Pinkovskiy on the evolution of poverty in Africa, and it contains exciting news: African poverty is falling and is falling rapidly since 1995 (this contradicts some older research). Moreover, this evolution is remarkably general across African countries, and not just explained by good news in a few large countries. Poverty is falling even in countries which are believed to burdened by geography, bad agricultural prospects, a history of slave trade, war, or lack of natural resources. And, to make the good news complete: income inequality has also decreased, and the Millennium Development Goal of halving the proportion of people earning less than $1 a day will be achieved on time.

You can see the reduction of the poverty rate in Africa in the graph below. From a “high point” of almost 45% of the population surviving on less than $1 dollar day in the late 1980s, that rate has fallen to 32% in 2006. How come? As you can also see in the graph, at the time poverty began to decline around 1995, GDP began to grow (after three decades of zero or negative growth). The graph shows a striking correlation between poverty reduction and economic growth, something I have written about before in another context, see here and here).

(source)

Of course, poverty reduction isn’t the automatic result of GDP growth only. Other factors are at work as well, but the paper is silent about those.

What’s interesting is that this African growth spurt since 1995 (probably briefly interrupted by the current recession) isn’t just caused by growing oil prices. If that had been the case, we would have seen increasing income inequality, since revenues from the oil industry are typically appropriated by elites. But that’s not the case. Poverty reduction has gone hand in hand with a reduction in income inequality. You can see the extent of this reduction in the following two graphs from the paper:

(source)

This means that growth has benefited the poor. However, although the reduction in poverty is impressive, it’s not quite as impressive as poverty reduction in China.

More on poverty measurement. More poverty statistics. More on Africa.

Lies, Damned Lies, and Statistics (27): Jumping to Conclusions, Ctd.

(source, cartoon by Eric Allie)

In a previous post in this series, I already mentioned the temptation to see things in data that just aren’t there, or to make data say things they don’t really say. I focused on the correlation-causation problem, a typical case of “jumping to conclusions”.

Elsewhere I gave the following example: there are data doing the rounds claiming that Republicans follow political news more closely than Democrats, which has some people saying that Republicans are more knowledgable and make better political choices. However, people don’t read more news because they are Republicans, but because they are relatively wealthy and older, and when they are they also tend to be more of the Republican type. So if you see data showing a correlation between political conservatism and attention to the news, don’t jump to conclusions and say that conservatives are inherently more attentive to the news, let alone that they make better political choices. A young and relatively poor conservative probably pays less attention than a wealthy and older liberal. Attention isn’t a function of political orientation. It has other causes.

However, as is evident from the cartoon above, data don’t have to be of the correlation type for people to see things in them that aren’t there. People have indeed interpreted popular rejection of healthcare reform or of the Obama administration in general as an expression of underlying racism, as if there can’t be any other reasons for rejection.*

Regarding the specific issue mentioned in the cartoon, there’s also another interesting statistical point related to the difficulty of doing a good survey (see also here, here and here):

Polling on the health-care bill is … complicated. Voters don’t know much about the plan. Most disapprove of it, but many disapprove because they want to see it go further. (source)

So there’s a “double jump” to conclusions in the cartoon:

• First, jumping from disapproval of healthcare reform to anti-Obama racism (blaming the former on the latter when this isn’t shown by the data), which is ridiculed, rightly to the extent that it is something real.
• Second, jumping from disapproval ratings on “something” to disapproval ratings on “healthcare reform”. The data only show that people disapprove of “something”: people may disapprove of only a part of healthcare reform, or may disapprove of the fact that it doesn’t go far enough rather than disapprove of reform as such; or they may disapprove of something that is not really proposed and hence misunderstand the whole thing and base their disapproval on lack of knowledge. Needless to say, this second jump in the cartoon is quite unconscious and probably not on purpose.

All this jumping is quite understandable. We always have to interpret data, and we can easily lose our way in the process. It’s also tempting to “find” explanations for data that fit with our pre-established opinions and biases.

* Personally, I’m in favor of reform.

More posts in this series are here. More on healthcare. More on racism.

Terrorism and Human Rights (26): Is Terrorism Caused by Unemployment?

girl holding bomb, by Bansky

(source, more Banksy here)

I don’t think I need to spell out the ways in which terrorism is a human rights issue (beyond the obvious violations of the human rights of the direct victims of terrorism there are serious human rights implications of the so-called ”war on terror“).

Some time ago, I linked to a paper claiming that poverty and lack of education do not, contrary to common belief, contribute to terrorism. If this claim is correct, then it has major implications for counter-terrorism efforts. There’s another paper here making a similar claim, looking at the correlation between violent insurgencies and levels of unemployment, specifically in Iraq and the Philippines. One often assumes that unemployment and the economic and social alienation resulting from it, are elements causing or facilitating political violence, and that efforts to promote employment can have a beneficial effect on social cohesion and political loyalty. The unemployed are believed to have the mindset (frustration etc.), the time and the opportunity to radicalize and be radicalized, whereas people who are employed have a lot to lose, economically, from political instability. Positively stated,

insurgency is a low-skill occupation so that creating jobs for the marginal unemployed reduces the pool of potential recruits.

However, the authors find

a robust negative correlation between unemployment and attacks against government and allied forces and no significant relationship between unemployment and the rate of insurgent attacks that kill civilians. … The negative correlation of unemployment with violence indicates that aid and development efforts that seek to enhance political stability through short-term job creation programs may well be misguided.

Some of the reasons given in the paper in order to explain this negative correlation are:

• Counter-insurgency forces usually spend money to buy intelligence from the general population. More unemployment means that the available money can buy more intelligence, hence bring levels of violence down.
• Insurgents also need to live. If there’s a lot of unemployment, they need to spend more time on basic survival and hence can spend less time on violence.
• Efforts to enhance security—establishing checkpoints and the like—damage the economy.
• etc.

The paper deals only with two countries, neither of which is perhaps a very typical case. Moreover, cross-border terrorism doesn’t seem to fit well into the analysis. But still, the findings are interesting.

The Causes of Poverty (30): Lack of Economic Growth

As stated in a previous post on the same subject, when a country achieves a certain level of economic growth – or, more precisely, rising levels of GDP per capita because economic growth as such can be the result of rising population levels – it is assumed that this reflects a higher average standard of living for its citizens. Economic growth is therefore seen as an important tool in the struggle against poverty (if you wonder why poverty is a human rights issue, go here). If a country is richer in general, the population will also be richer on average. On average meaning that GDP growth isn’t necessarily equally distributed over every member of the population. That is why GDP growth isn’t sufficient proof of poverty reduction. Separate measurements of poverty and inequality are necessary.

So in theory, you can have GDP growth and increasing levels of poverty, on the condition that GDP growth is concentrated in the hands of a few. However, that’s generally not the case. GDP growth benefits to some extent many of the poor as well as the wealthy, which is shown by the strong correlation between poverty reduction and levels of GDP growth (always per capita of course). It’s no coincidence that a country such as China, which has seen strong GDP growth over the last decades, is also a country that has managed to reduce poverty levels substantially.

Unfortunately, growth isn’t a silver bullet. Poverty is a complex problem, requiring many types of solutions. Promoting economic growth will do a lot of the work, but something more is required. In a new paper, Martin Ravaillon gives the example of China, Brazil and India. The levels of poverty reduction in these three countries, although impressive, do not simply mirror the levels of economic growth. Although half of the world’s poor live in these three countries, in the last 25 years China has reduced its poverty level from 84% of the population in 1981 to just 16% in 2005 (see chart below). China is exceptional, but Brazil also did well, cutting its rate in half over the same period (8% of Brazilians still live on less than $1.25 a day). Regarding India, there are some problems with its statistics, but whichever statistic you use, there’s a clear reduction.

poverty reduction in China, India and Brazil

(source)

Ravaillon points out that the intensity of poverty reduction was higher in Brazil than in India and China, despite lower GDP growth rates.

Per unit of growth, Brazil reduced its proportional poverty rate five times more than China or India did. How did it do so well? The main explanation has to do with inequality. This (as measured by the Gini index, also marked on the chart) has fallen sharply in Brazil since 1993, while it has soared in China and risen in India. Greater inequality dampens the poverty-reducing effect of growth. (source)

Which is rather obvious: higher levels of income equality means a better distribution of the benefits of growth. So the “pro-growth strategy” against poverty is important but not enough, and should be combined with Brazilian type anti-inequality measures (focus on education, healthcare and redistribution).

Why Do Countries Become/Remain Democracies? Or Don’t? (7): Education and Prosperity

(This is a follow-up from a previous post).

There aren’t many questions in political science that are more important than this one: which are the factors that determine whether a country becomes or doesn’t become a democracy, and that determine the degree to which a country is democratic. There are two reasons why this question is important:

• Democracy is an important good. I gave some arguments for this statement here. Hence it’s important to know what facilitates or hinders the realization of this good.
• Countries act on this statement in their foreign policy. For example, part of the rationale for invading Iraq was the conviction held by the U.S. administration of the time that promoting democracy in Iraq was both an intrinsic good and in the interest of the U.S. (see here as well).

I gave a short and non-exhaustive list of possible factors promoting/undermining the development/survival of democracy here. In the current post I want to focus on two of them: education levels and income or prosperity levels.

1. Education

education and democracy correlation

(source)

This graph compares the Polity IV Democracy Index scores for the countries of the world (average scores during the 1960-2000 period), with the average years of schooling of the adult population in 1960. And there’s obviously a correlation, and the quote below gives an indication about the direction of correlation:

The chart above shows the 77 percent correlation between education levels in 1960 (measured by the average years of schooling in a country as estimated by Robert Barro and Jong-Wha Lee), and the subsequent 40-year average of the Polity IV democracy index. That democracy index runs from zero to 10, where countries with index values less than three don’t look remotely democratic and countries with index values of about seven are reasonably well-functioning democracies.

One way to read the graph is that there are basically no countries with very low levels of education that have managed to be democratic over the long term, and almost every country with a high level of education has remained a stable democracy.

Thomas Jefferson wrote that “if a nation expects to be ignorant and free, in a state of civilization, it expects what never was and never will be.” In 1960, 36 nations had less than 1.74 years of schooling (which happens to be the level that Afghanistan has today). Of those 36 countries, only two — India and Botswana — managed to have average democracy scores above 4.2.

Out of the 19 countries in this sample with more than 5.3 years of schooling (the current level in Iran) in 1960, 17 have average democracy scores above 7.9. Fifteen of these have been perfectly democratic, at least by the standards of Polity IV. Only Poland and Hungary were dictatorships, and one can certainly argue that those places would have been democracies in 1960s if it were not for Soviet troops.

But in the middle ranges of education, between two and five years on average, almost anything goes. Some places, like Costa Rica and Italy, have been extremely democratic, while others, like Kuwait and Paraguay, have not. Iraq falls into this category today, which suggests a fair amount of uncertainty about that country’s political future.

Why do I think that the chain of causality runs from education to democracy rather than the reverse? Democracy in 1960 is essentially uncorrelated with subsequent growth in the levels of education. Education in 1960, on the other hand, does an extremely good job of predicting increases in democracy.

Why is there a connection between human capital and freedom? Giacomo Ponzetto, Andrei Shleifer and I have argued that the connection reflects the ability of educated people to organize and fight collaboratively. Dictators provide strong incentives for the ruling clique; democracies provide more modest benefits for everyone else. For democracy to beat dictatorship, the dispersed population needs to have the skills and motivation to work collaboratively to defeat dictatorial coups and executive aggrandizement.

Education teaches skills, like reading and writing, that enable people to work collaboratively. At younger grades, teachers spend a lot of time teaching children how to get along. In the United States, education is strongly linked to civic engagement and membership in social groups. The ability to work together enables the defense of democracy. Edward L. Glaeser (source)

2. Income

There’s an interesting paper here examining the causal relation between democracy and income. The authors find that

the level of national income provides the most important factor explaining inter-country variations in the degree of democracy with the consequence that low income is the most important barrier to democracy.

They first present the correlation between income and democracy, using not the Polity IV index but the Gastil/Freedom House index (see also here):

democracy and income correlation

(source)

The authors have two reasons to believe that the causal link goes from income to democracy rather than the other way around:

• Initial income in 1971 correlates with average democracy scores during the 1972-2005 period. This approach is similar to the one above in the case of education and democracy.
• And – simultaneously – there doesn’t seem to be a very strong causal link going from democracy to income. Barro has concluded that the degree of democracy is only a minor variable explaining income levels. So there is only a weak causal link going from democracy to income (see also here*). This means that the strong correlation shown in the graph above must be explained by a causal link going from income to democracy.

Why do higher levels of income promote the development of democracy? I gave an overview of the reasons here but some of the more important ones are:

• Higher education levels in a population means a higher probability of contestation. Following the Maslow hierarchy of needs it’s natural to expect the appearance of political needs once more basic needs have been secured.
• More income means more complex production. This in turn means that governments find it harder to impose central control over their economies.

Obviously, income is just one of many factors determining the development of democracy. It’s an important one, but clearly not sufficient. The graph above shows the Muslim countries separately. As you can see, all non-Muslim countries with high income levels are in the “high level of democracy” range. Affluent Muslim countries, however, aren’t. This indicates that affluence in itself promotes but doesn’t determine the development of democracy. Other factors are also in play. Culture and religion are perhaps some of them. It’s often argued that Islam is incompatible with democracy, or at least slows down the development of or transition to democracy. I’ll come back to this controversial topic another time.

* One can argue that the link would be stronger if democracies would be of better quality, see here.

Lies, Damned Lies, and Statistics (24): Mistakes in the Direction of Causation

Time for a more lighthearted post in this blog series. Suppose you find a correlation between two phenomena. And you’re tempted to conclude that there’s a causal relation as well. The problem is that this causal relation – if it exists at all – can go either way. It’s a common mistake – or a case of fraud, as it happens - to choose one direction of causation and forget that the real causal link can go the other way, or both ways at the same time.

An example. We often think that people who play violent video games are more likely to show violent behavior because they are incited by the games to copy the violence in real life. But can it not be that people who are more prone to violence are more fond of violent video games? (See also here). We choose a direction of causation that fits with our pre-existing beliefs.

Another widely shared belief is that uninformed and uneducated voters will destroy democracy, or at least diminish its value (see here, here and here). No one seems to ask the question whether it’s not a diminished form of democracy that renders citizens apathetic and uninformed. Maybe a full or deep democracy can encourage citizens to participate and become more knowledgeable through participation.

A classic example is the correlation between education levels and GDP (see also here). Do countries with higher education levels experience more economic growth because of the education levels of their citizens? Or is it that richer countries can afford to spend more on education and hence have better educated citizens? Probably both.

Another cartoon that expresses the same risk:

(source)

More posts in this blog series.

Statistical Jokes (2): Fun With Correlation, Ctd.

(if I understand this one correctly: “it’s not because you always see me in the neighborhood of a mess that I caused the mess”; fair enough)

More “textual” information on the correlation-causation problem is here. More statistical jokes are here.

Statistical Jokes (1): Fun With Correlation

The Japanese eat very little fat and suffer fewer heart attacks than the British or the Americans.

On the other hand, the French eat a lot of fat and also suffer fewer heart attacks than the British or the Americans.

The Japanese drink very little red wine and suffer fewer heart attacks than the British or the Americans.

The Italians drink excessive amounts of red wine and also suffer fewer heart attacks than the British or the Americans.

Conclusion: Eat and drink whatever you like. It’s speaking English that kills you. (source)

Lies, Damned Lies, and Statistics (17): The Correlation-Causation Problem and Omitted Variable Bias, aka “Jumping to Conclusions”

correlation vs causation

(source)

Some more detailed information after my casual remark on the correlation-causation problem. Here’s a fictitious example of what is meant by “Omitted Variable Bias“, a type of statistical bias that illustrates this problem. Suppose we see from Department of Defense data that male U.S. soldiers are more likely to be killed in action than female soldiers. Or, more precisely and in order to avoid another statistical error, the percentage of male soldiers killed in action is larger than the percentage of female soldiers. So there is a correlation between the gender of soldiers and the likelihood of being killed in action.

One could – and one often does – conclude from such a finding that there is a causation of some kind: the gender of soldiers increases the chances of being killed in action. Again more precisely: one can conclude that some aspects of gender – e.g. a male propensity for risk taking – leads to higher mortality.

However, it’s here that the Omitted Variable Bias pops up. The real cause of the discrepancy between male and female combat mortality may not be gender or a gender related thing, but a third element, an “omitted variable” which doesn’t show in the correlation. In our fictional example, it may be the type of deployment: it may be that male soldiers are more commonly deployed in dangerous combat operations, whereas female soldiers may be more active in support operations away from the front-line.

correlation and causation

(source)

OK, time for a real example. It has to do with home-schooling. In the U.S., many parents decide to keep their children away from school and teach them at home. For different reasons: ideological ones, reasons that have to do with their children’s special needs etc. The reasons are not important here. What is important is that many people think that home-schooled children are somehow less well educated (parents, after all, aren’t trained teachers). However, proponents of home-schooling point to a study that found that these children score above average in tests. However, this is a correlation, not necessarily a causal link. It doesn’t prove that home-schooling is superior to traditional schooling. Parents who teach their children at home are, by definition, heavily involved in their children’s education. The children of such parents do above average in normal schooling as well. The omitted variable here is parents’ involvement. It’s not the fact that the children are schooled at home that explains their above average scores. It’s the type of parents. Instead of comparing home-schooled children to all other children, one should compare them to children from similar families in the traditional system.

(source)

Greg Mankiw believes he has found another example of Omitted Variable Bias in this graph plotting test scores for U.S. students against their family income:

(source, the R-square for each test average/income range chart is about 0.95)

[T]he above graph … show[s] that kids from higher income families get higher average SAT scores. Of course! But so what? This fact tells us nothing about the causal impact of income on test scores. … This graph is a good example of omitted variable bias … The key omitted variable here is parents’ IQ. Smart parents make more money and pass those good genes on to their offspring. Suppose we were to graph average SAT scores by the number of bathrooms a student has in his or her family home. That curve would also likely slope upward. (After all, people with more money buy larger homes with more bathrooms.) But it would be a mistake to conclude that installing an extra toilet raises yours kids’ SAT scores. … It would be interesting to see the above graph reproduced for adopted children only. I bet that the curve would be a lot flatter. Greg Mankiw (source)

Meaning that adopted children, who usually don’t receive their genes from their new families, have equal test scores, no matter if they have been adopted by rich or poor families. Meaning in turn that the wealth of the family in which you are raised doesn’t influence your education level, test scores or intelligence.

However, in his typical hurry to discard all possible negative effects of poverty, Mankiw may have gone a bit too fast. While it’s not impossible that the correlation is fully explained by differences in parental IQ, other evidence points elsewhere. I’m always suspicious of theories that take one cause, exclude every other type of explanation and end up with a fully deterministic system, especially if the one cause that is selected is DNA. Life is more complex than that. Regarding this particular matter, take a look back at this post, which shows that education levels are to some extent determined by parental income (university enrollment is determined both by test scores and by parental income, even to the extent that people from high income families but with average test scores, are slightly more likely to enroll in university than people from poor families but with high test scores).

What Mankiw did, in trying to avoid the Omitted Variable Bias, was in fact another type of bias, one which we could call the Singular Variable Bias: assuming that a phenomenon has a singular cause. In honor of Professor Mankiw (who does some good work, see here for example), I propose that henceforth we call it the Mankiw Bias.

More posts in this series.

The Ethics of Human Rights (21): Social Mobility, Egalitarianism, Equality of Opportunity, and Meritocracy

equality of opportunity

(source, cartoon by Dario Castillejos)

In the best egalitarian society, people can change occupations, groups, associations etc. but their income, poverty level or social class will not change a lot as a result of this, since there’s not much difference between different income levels. This means that the society in question has decided that different occupations, talents and efforts should receive roughly the same financial reward. That may or may not be a good thing. Intuitively I would say that some occupations and some amounts of effort investment should receive higher financial rewards than others, in which case a somewhat inegalitarian society is what I want, notwithstanding my concerns about the problems created by inequality (see here for example). What I certainly don’t want is the worst egalitarian society, which combines the problem of equal rewards for morally diverse activities with the problem of fixed occupations and lack of social mobility, Soviet style.

In the worst inegalitarian society, there isn’t a lot of social mobility, social mobility in the sense of children ending up in adult life in a higher or lower level of income than the level of their parents.* There may be relatively many people changing occupations, but always within a limited class of occupations that yield roughly the same income levels. Such a lack of social mobility is an indication that income levels are not the result of merit, desert, reward, effort or talent, but rather the result of society’s choice not to equalize opportunities and to let people’s opportunities be determined by factors such as the family in which they happen to be born, unequal access to education etc. Genes do play a role in determining talent, and perhaps even willingness to invest effort, but only if genes were the sole force determining talent and effort could we claim that a lack of social mobility in an inegalitarian society is an inevitable characteristic of this society and not the consequence of a conscious choice of this society.

Since I don’t believe that genes have such a strong determining force, I have to conclude that the worst inegalitarian society chooses to limit social mobility and to accept (or even promote) unequal opportunities. Such a society in fact chooses to be a class society, a society that limits entry and exit into the various classes or income level groups and that forces parents and their adult children to share similar income levels (income levels are transmitted across generations).

The limited power of genes also allows me to conclude, positively now, that the best inegalitarian society can and should try to enact policies that promote social mobility. Such policies should remove obstacles that hinder people from using their talents and efforts in order to achieve a position in society that corresponds to a higher income level than the level their parents “enjoy”. These obstacles can be parental poverty, lack of access to quality education or to cultural resources, parental crime, peer pressure etc. In short, the best inegalitarian society should try to equalize opportunities. People with similar talents and willingness to develop and use these talents should have a roughly equal chance of ending up in a similar income level. If they don’t have such an equal chance, then it means that they don’t have the same opportunities and that certain obstacles hinder some of these people in the use and development of their talents. I can see no reason why the imposition of such obstacles on some people and not on others could ever be justified, but I’m open to suggestions.

Those who are at the same level of talent and ability, and have the same willingness to use them, should have the same prospects of success regardless of their initial place in the social system. In all sectors of society there should be roughly equal prospects of culture and achievement for everyone similarly motivated and endowed. The expectations of those with the same abilities and aspirations should not be affected by their social class. Chances to acquire cultural knowledge and skills should not depend upon one’s class position, and so the school system, whether public or private, should be designed to even out class barriers. John Rawls (source)

John Rawls

If we assume that genes have a limited role in distributing talent, that the distribution of talent among people is therefore to some extent random and not determined by who their parents are; and if we further assume that the willingness to invest effort isn’t completely determined by parental influence or by genetics – and if, on top of that, opportunities are equalized (to some extent), then we should find a lack of correlation between the economic status of parents and their children. We should, in other words, find high levels of social mobility. If not, the influence of genes on talent and the influence of parents on the willingness to invest effort are more powerful than we think; or – more likely – the society hasn’t been successful in creating equality of opportunity (hasn’t provided equal access to quality education for instance). The levels of mobility are therefore a good indicator of the equality of opportunity in a society.

If the best inegalitarian society tries to equalize opportunities and is reasonably successful, then this doesn’t mean that it will necessarily become an egalitarian society. Equalizing opportunities doesn’t imply equalizing rewards for different activities, and neither does it mean that everyone will make equally successful use of the equal opportunities. There will be a lot of social mobility and a lack of correlation between the social position of parents and children, but the mobility can go up for some people and down for others, depending on the talents people have, the efforts they are willing to invest, and the rewards that society gives to particular talents, activities and efforts. Because of these different rewards, and because equal opportunities will be used unequally, there is no reason to expect a convergence of income levels. The best inegalitarian society will become a meritocracy, which produces, by definition, unequal income levels because it differentiates between deserving and less deserving activities, and between deserving and less deserving efforts within an activity.

This kind of society differs fundamentally from the worst inegalitarian society which is a class society and which therefore locks people in positions whatever their merits (class society can mean different things – caste society, nepotistic society etc. – but the effect is always the same). It also differs from the best egalitarian society which allows people to move between occupations but rewards all occupations equally and can’t therefore be called a meritocracy.

I mentioned before that a society can choose to be the best or the worst inegalitarian society. But how does it do that? “Society” is a vague concept. Who are the people actually making those choices? Well, it can be the politicians for instance. It’s quite clear that different policies have different effects on the equality of opportunities and on social mobility. Estate taxes or inheritance taxes play a huge role. Redistribution policies and policies aimed at education as well. But the processes leading towards and away from equality of opportunity can also be more below the surface:

It turns out that there’s a bit of a paradoxical relationship between believing your country has a lot of economic mobility and your country actually having a lot of economic mobility. If you believe that your country is extremely mobile, you’re likely to believe the results of the economic competition are relatively fair. As such, you won’t want to slap the rich with particularly high tax rates and you won’t be terribly concerned about spreading economic opportunity. After all, anyone can make it!

On the other hand, if you don’t believe your country is terribly mobile, then you’re less likely to believe economic outcomes are fair. And if you don’t believe the outcomes are fair, you’re likely to tax the winners relatively heavily and plow those profits into things like universal health care and free college. Policies, in other words, that spread opportunity more widely and thus make your society more mobile. Put like that, it sort of makes sense. If you believe your society is already economically mobile, you don’t spend a lot of time trying to solve the problem of insufficient economic mobility. if you don’t believe that, then you implement policies meant to increase mobility. Ezra Klein (source)

Some data on social mobility are here, here, here, and here. And there’s something interesting about what’s called a Human Opportunity Index here. More posts in this series are here.

* “Basically social mobility refers to the likelihood that a child will grow up into adulthood and attain a higher level of economic and social wellbeing than his/her family of origin. Is there a correlation between the socioeconomic status (SES) of an adult and his/her family of origin? Do poor people tend to have poor parents? And do poor parents tend to have children who end up as poor adults later in life? Does low SES in the parents’ circumstances at a certain time in life – say, the age of 30 – serve to predict the SES of the child at the same age?” (source)

Economic Human Rights (30): Life Expectancy Throughout History

How is life expectancy relevant for human rights? High levels of life expectancy can mean a long life of oppression and cruelty, but it’s fair to say that a long life is generally beneficial for human rights, and that low average life expectancy rates are indicators of human rights violations. The longer people life, on average, the more they can do with their lives, and the more they can enjoy their freedom. If people’s lives are shorter, on average, it’s likely that this is because of human rights violations. For example, because:

• their healthcare isn’t provided for (and healthcare is a human right)
• their poverty traps them in a vicious circle of poverty and ill health (and poverty is a human right)
• their governments send them to war, etc.

So it’s useful to note that life expectancy, over the course of human history, has risen sharply, especially during modern times:

life expectancy through the ages

(source, click on the image to enlarge)

Life expectancy during much of pre-modern history averaged just below 30 years. Part of the reason for such a low figure is that many children died at a very young age, pulling down the average life expectancy. Those who didn’t die young had a good chance of surviving to what we now call “middle age”.

After the Industrial Revolution many more children survived into adulthood and by the beginning of the 20th century average life expectancy in the developed world was close to 50, whereas for the world as a whole it was only around 40 years. The figures now are 78 and 67 respectively. This graph shows the rapid and sudden improvement after centuries of stagnation:

Life Expectancy throughout history, long trend

(source)

The reason for this sudden improvement during and after the industrial revolution is a combination of improved medical technology and higher wealth. Not surprisingly, life expectancy is highly correlated with income levels – more wealth means higher investment in healthcare, less war etc. – but not in a linear fashion: the U.S. has very high GDP per capita but not higher life expectancy than some countries/regions with somewhat lower income levels (some blame the healthcare system, others the life-style choices of many Americans). And, compared to Africa, India has higher life expectancy with similar income levels (the HIV/AIDS epidemic is part of the explanation).

(source)

There’s a map comparing life expectancy in the world here. And there are some more statistics on life expectancy here.

Human Rights Cartoon (61): Corruption

(source, cartoon by Angel Boligan)

Corruption, or “the misuse of public office for private gain”, is not a human rights violation as such (there is no right not to suffer the consequences of corruption), but it is the cause of various rights violations. Notably, it has an impact on economic growth (see here) and hence also on poverty reduction (given the correlation between growth and poverty reduction,  see here). Corruption also has an impact on poverty on the level of individuals rather than countries (and there is a right not to suffer poverty). It’s obvious that individuals can make better use of the funds that they (have to) spend on bribes. As depicted in the cartoon, those that are forced to pay bribes are often people who are already vulnerable.

Moreover, corruption eats away at the rule of law. Even in the most corrupt countries, corruption is usually illegal. If illegal activity becomes normal practice, the rule of law is obviously undermined, with possible consequences for judicial protection in general, including protection of human rights. Even more seriously, corruption is associated with political instability since it tends to reduce citizens’ trust and faith in institutions.

Some statistics on corruption are here, here, here, and here.

Measuring Democracy (4): How, Why and What

(source, photo by Samantha Nicol)

How do we measure democracy?

How do we assess if a country is a democracy or not, or is more democratic than another? Or, in other words, how do we assess the “democraticness” of a country, or the level or quality of its democracy, if any? It’s obvious from this way of phrasing the question that my preferred system of measurement will not be binary or dichotomous. I want to have a measurement system that gives me more than merely an indication of the presence or absence of democracy in a country. I want a scale of “democraticness”, ranging from total absence of any elements of democracy to a perfect democracy, with as many intermediate levels as possible. How this can be done is another matter. Different people have tried to do this in different ways, none of them in a satisfactory way. (Even the one with the best reputation can be criticized). So someone should come up with something new.

Why should we measure democracy?

But apart from how, there’s the question of why. Why would we want to measure the quality of democracy? For two reasons I think.

1. Democracy is said to have a beneficial influence on peace, prosperity, equality and freedom. But the correct measurement of this influence depends on the accuracy of democracy measurement. If we can’t measure democracy, we can’t measure how it is correlated with other measures.
2. Secondly, democracy is valuable in itself. Ideally, it allows citizens to participate in the process that shapes their collective fate. Self-government, self-determination and self-control are aspects of freedom. To the extent that freedom is considered to be valuable, democracy is as well. A better measure of democracy makes it easier to progress towards a better democracy, and hence towards more freedom.

What is it that we want to measure?

This answer to the question “why measure democracy?” hints at another important question: “what is it that we want to measure?”. Democracy is a highly contested concept, and no two people mean exactly the same when they use the word. That doesn’t mean we can’t measure democracy. It simply means that anyone proposing a measurement system – as I won’t do here obviously – must make it perfectly clear what he or she personally and controversially means by the concept. Those who believe that this meaning is nonsense or contains mistakes about the theory of democracy can reject the measurement system.

I already mentioned that my preferred concept of democracy to be used in measurement systems is an “ideal democracy” or a “full democracy”. So the point is not to attempt to measure something which exists in reality. The measurement should attempt to calculate the distance between existing political regimes and the ideal of democracy, as one sees it. It may appear that such a choice makes things even more controversial, and increases the likelihood that a measurement project will be rejected by large numbers of readers who, after all, have probably different views on the nature of an ideal democracy. It’s more likely that there exist different points of view on ideals than on facts. However, the advantage of measuring the distance between actual regimes and an ideal, compared to measuring how many countries share the same, existing regime type, is that the targets of the measurement will be more positive and accepting about it. After all, it comes over better if you are somewhere on a continuum towards full democracy and if you’re being compared to an ideal, than if you are simply classified as a non-democracy/oligarchy/theocracy etc. and compared to a certain country.

More posts in this series.

Lies, Damned Lies, and Statistics (1): Correlation

You know I love graphs and statistics, so here’s one showing how importing lemons from Mexico reduces highway fatality rates in the U.S.:

(source)

And here‘s another one. Just so that you don’t automatically believe everything I write (as if you would), and a funny reminder that correlation doesn’t necessarily imply causation.

For some real statistics, see here. For something more on the famous quote in the title, go here.